1config ARCH 2 string 3 option env="ARCH" 4 5config KERNELVERSION 6 string 7 option env="KERNELVERSION" 8 9config DEFCONFIG_LIST 10 string 11 depends on !UML 12 option defconfig_list 13 default "/lib/modules/$UNAME_RELEASE/.config" 14 default "/etc/kernel-config" 15 default "/boot/config-$UNAME_RELEASE" 16 default "$ARCH_DEFCONFIG" 17 default "arch/$ARCH/defconfig" 18 19config CONSTRUCTORS 20 bool 21 depends on !UML 22 23config IRQ_WORK 24 bool 25 26config BUILDTIME_EXTABLE_SORT 27 bool 28 29menu "General setup" 30 31config BROKEN 32 bool 33 34config BROKEN_ON_SMP 35 bool 36 depends on BROKEN || !SMP 37 default y 38 39config INIT_ENV_ARG_LIMIT 40 int 41 default 32 if !UML 42 default 128 if UML 43 help 44 Maximum of each of the number of arguments and environment 45 variables passed to init from the kernel command line. 46 47 48config CROSS_COMPILE 49 string "Cross-compiler tool prefix" 50 help 51 Same as running 'make CROSS_COMPILE=prefix-' but stored for 52 default make runs in this kernel build directory. You don't 53 need to set this unless you want the configured kernel build 54 directory to select the cross-compiler automatically. 55 56config COMPILE_TEST 57 bool "Compile also drivers which will not load" 58 default n 59 help 60 Some drivers can be compiled on a different platform than they are 61 intended to be run on. Despite they cannot be loaded there (or even 62 when they load they cannot be used due to missing HW support), 63 developers still, opposing to distributors, might want to build such 64 drivers to compile-test them. 65 66 If you are a developer and want to build everything available, say Y 67 here. If you are a user/distributor, say N here to exclude useless 68 drivers to be distributed. 69 70config LOCALVERSION 71 string "Local version - append to kernel release" 72 help 73 Append an extra string to the end of your kernel version. 74 This will show up when you type uname, for example. 75 The string you set here will be appended after the contents of 76 any files with a filename matching localversion* in your 77 object and source tree, in that order. Your total string can 78 be a maximum of 64 characters. 79 80config LOCALVERSION_AUTO 81 bool "Automatically append version information to the version string" 82 default y 83 help 84 This will try to automatically determine if the current tree is a 85 release tree by looking for git tags that belong to the current 86 top of tree revision. 87 88 A string of the format -gxxxxxxxx will be added to the localversion 89 if a git-based tree is found. The string generated by this will be 90 appended after any matching localversion* files, and after the value 91 set in CONFIG_LOCALVERSION. 92 93 (The actual string used here is the first eight characters produced 94 by running the command: 95 96 $ git rev-parse --verify HEAD 97 98 which is done within the script "scripts/setlocalversion".) 99 100config HAVE_KERNEL_GZIP 101 bool 102 103config HAVE_KERNEL_BZIP2 104 bool 105 106config HAVE_KERNEL_LZMA 107 bool 108 109config HAVE_KERNEL_XZ 110 bool 111 112config HAVE_KERNEL_LZO 113 bool 114 115config HAVE_KERNEL_LZ4 116 bool 117 118choice 119 prompt "Kernel compression mode" 120 default KERNEL_GZIP 121 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 122 help 123 The linux kernel is a kind of self-extracting executable. 124 Several compression algorithms are available, which differ 125 in efficiency, compression and decompression speed. 126 Compression speed is only relevant when building a kernel. 127 Decompression speed is relevant at each boot. 128 129 If you have any problems with bzip2 or lzma compressed 130 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older 131 version of this functionality (bzip2 only), for 2.4, was 132 supplied by Christian Ludwig) 133 134 High compression options are mostly useful for users, who 135 are low on disk space (embedded systems), but for whom ram 136 size matters less. 137 138 If in doubt, select 'gzip' 139 140config KERNEL_GZIP 141 bool "Gzip" 142 depends on HAVE_KERNEL_GZIP 143 help 144 The old and tried gzip compression. It provides a good balance 145 between compression ratio and decompression speed. 146 147config KERNEL_BZIP2 148 bool "Bzip2" 149 depends on HAVE_KERNEL_BZIP2 150 help 151 Its compression ratio and speed is intermediate. 152 Decompression speed is slowest among the choices. The kernel 153 size is about 10% smaller with bzip2, in comparison to gzip. 154 Bzip2 uses a large amount of memory. For modern kernels you 155 will need at least 8MB RAM or more for booting. 156 157config KERNEL_LZMA 158 bool "LZMA" 159 depends on HAVE_KERNEL_LZMA 160 help 161 This compression algorithm's ratio is best. Decompression speed 162 is between gzip and bzip2. Compression is slowest. 163 The kernel size is about 33% smaller with LZMA in comparison to gzip. 164 165config KERNEL_XZ 166 bool "XZ" 167 depends on HAVE_KERNEL_XZ 168 help 169 XZ uses the LZMA2 algorithm and instruction set specific 170 BCJ filters which can improve compression ratio of executable 171 code. The size of the kernel is about 30% smaller with XZ in 172 comparison to gzip. On architectures for which there is a BCJ 173 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ 174 will create a few percent smaller kernel than plain LZMA. 175 176 The speed is about the same as with LZMA: The decompression 177 speed of XZ is better than that of bzip2 but worse than gzip 178 and LZO. Compression is slow. 179 180config KERNEL_LZO 181 bool "LZO" 182 depends on HAVE_KERNEL_LZO 183 help 184 Its compression ratio is the poorest among the choices. The kernel 185 size is about 10% bigger than gzip; however its speed 186 (both compression and decompression) is the fastest. 187 188config KERNEL_LZ4 189 bool "LZ4" 190 depends on HAVE_KERNEL_LZ4 191 help 192 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding. 193 A preliminary version of LZ4 de/compression tool is available at 194 <https://code.google.com/p/lz4/>. 195 196 Its compression ratio is worse than LZO. The size of the kernel 197 is about 8% bigger than LZO. But the decompression speed is 198 faster than LZO. 199 200endchoice 201 202config DEFAULT_HOSTNAME 203 string "Default hostname" 204 default "(none)" 205 help 206 This option determines the default system hostname before userspace 207 calls sethostname(2). The kernel traditionally uses "(none)" here, 208 but you may wish to use a different default here to make a minimal 209 system more usable with less configuration. 210 211config SWAP 212 bool "Support for paging of anonymous memory (swap)" 213 depends on MMU && BLOCK 214 default y 215 help 216 This option allows you to choose whether you want to have support 217 for so called swap devices or swap files in your kernel that are 218 used to provide more virtual memory than the actual RAM present 219 in your computer. If unsure say Y. 220 221config SYSVIPC 222 bool "System V IPC" 223 ---help--- 224 Inter Process Communication is a suite of library functions and 225 system calls which let processes (running programs) synchronize and 226 exchange information. It is generally considered to be a good thing, 227 and some programs won't run unless you say Y here. In particular, if 228 you want to run the DOS emulator dosemu under Linux (read the 229 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>), 230 you'll need to say Y here. 231 232 You can find documentation about IPC with "info ipc" and also in 233 section 6.4 of the Linux Programmer's Guide, available from 234 <http://www.tldp.org/guides.html>. 235 236config SYSVIPC_SYSCTL 237 bool 238 depends on SYSVIPC 239 depends on SYSCTL 240 default y 241 242config POSIX_MQUEUE 243 bool "POSIX Message Queues" 244 depends on NET 245 ---help--- 246 POSIX variant of message queues is a part of IPC. In POSIX message 247 queues every message has a priority which decides about succession 248 of receiving it by a process. If you want to compile and run 249 programs written e.g. for Solaris with use of its POSIX message 250 queues (functions mq_*) say Y here. 251 252 POSIX message queues are visible as a filesystem called 'mqueue' 253 and can be mounted somewhere if you want to do filesystem 254 operations on message queues. 255 256 If unsure, say Y. 257 258config POSIX_MQUEUE_SYSCTL 259 bool 260 depends on POSIX_MQUEUE 261 depends on SYSCTL 262 default y 263 264config CROSS_MEMORY_ATTACH 265 bool "Enable process_vm_readv/writev syscalls" 266 depends on MMU 267 default y 268 help 269 Enabling this option adds the system calls process_vm_readv and 270 process_vm_writev which allow a process with the correct privileges 271 to directly read from or write to another process' address space. 272 See the man page for more details. 273 274config FHANDLE 275 bool "open by fhandle syscalls" 276 select EXPORTFS 277 help 278 If you say Y here, a user level program will be able to map 279 file names to handle and then later use the handle for 280 different file system operations. This is useful in implementing 281 userspace file servers, which now track files using handles instead 282 of names. The handle would remain the same even if file names 283 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2) 284 syscalls. 285 286config USELIB 287 bool "uselib syscall" 288 default y 289 help 290 This option enables the uselib syscall, a system call used in the 291 dynamic linker from libc5 and earlier. glibc does not use this 292 system call. If you intend to run programs built on libc5 or 293 earlier, you may need to enable this syscall. Current systems 294 running glibc can safely disable this. 295 296config AUDIT 297 bool "Auditing support" 298 depends on NET 299 help 300 Enable auditing infrastructure that can be used with another 301 kernel subsystem, such as SELinux (which requires this for 302 logging of avc messages output). Does not do system-call 303 auditing without CONFIG_AUDITSYSCALL. 304 305config HAVE_ARCH_AUDITSYSCALL 306 bool 307 308config AUDITSYSCALL 309 bool "Enable system-call auditing support" 310 depends on AUDIT && HAVE_ARCH_AUDITSYSCALL 311 default y if SECURITY_SELINUX 312 help 313 Enable low-overhead system-call auditing infrastructure that 314 can be used independently or with another kernel subsystem, 315 such as SELinux. 316 317config AUDIT_WATCH 318 def_bool y 319 depends on AUDITSYSCALL 320 select FSNOTIFY 321 322config AUDIT_TREE 323 def_bool y 324 depends on AUDITSYSCALL 325 select FSNOTIFY 326 327source "kernel/irq/Kconfig" 328source "kernel/time/Kconfig" 329 330menu "CPU/Task time and stats accounting" 331 332config VIRT_CPU_ACCOUNTING 333 bool 334 335choice 336 prompt "Cputime accounting" 337 default TICK_CPU_ACCOUNTING if !PPC64 338 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64 339 340# Kind of a stub config for the pure tick based cputime accounting 341config TICK_CPU_ACCOUNTING 342 bool "Simple tick based cputime accounting" 343 depends on !S390 && !NO_HZ_FULL 344 help 345 This is the basic tick based cputime accounting that maintains 346 statistics about user, system and idle time spent on per jiffies 347 granularity. 348 349 If unsure, say Y. 350 351config VIRT_CPU_ACCOUNTING_NATIVE 352 bool "Deterministic task and CPU time accounting" 353 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL 354 select VIRT_CPU_ACCOUNTING 355 help 356 Select this option to enable more accurate task and CPU time 357 accounting. This is done by reading a CPU counter on each 358 kernel entry and exit and on transitions within the kernel 359 between system, softirq and hardirq state, so there is a 360 small performance impact. In the case of s390 or IBM POWER > 5, 361 this also enables accounting of stolen time on logically-partitioned 362 systems. 363 364config VIRT_CPU_ACCOUNTING_GEN 365 bool "Full dynticks CPU time accounting" 366 depends on HAVE_CONTEXT_TRACKING 367 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN 368 select VIRT_CPU_ACCOUNTING 369 select CONTEXT_TRACKING 370 help 371 Select this option to enable task and CPU time accounting on full 372 dynticks systems. This accounting is implemented by watching every 373 kernel-user boundaries using the context tracking subsystem. 374 The accounting is thus performed at the expense of some significant 375 overhead. 376 377 For now this is only useful if you are working on the full 378 dynticks subsystem development. 379 380 If unsure, say N. 381 382config IRQ_TIME_ACCOUNTING 383 bool "Fine granularity task level IRQ time accounting" 384 depends on HAVE_IRQ_TIME_ACCOUNTING && !NO_HZ_FULL 385 help 386 Select this option to enable fine granularity task irq time 387 accounting. This is done by reading a timestamp on each 388 transitions between softirq and hardirq state, so there can be a 389 small performance impact. 390 391 If in doubt, say N here. 392 393endchoice 394 395config BSD_PROCESS_ACCT 396 bool "BSD Process Accounting" 397 help 398 If you say Y here, a user level program will be able to instruct the 399 kernel (via a special system call) to write process accounting 400 information to a file: whenever a process exits, information about 401 that process will be appended to the file by the kernel. The 402 information includes things such as creation time, owning user, 403 command name, memory usage, controlling terminal etc. (the complete 404 list is in the struct acct in <file:include/linux/acct.h>). It is 405 up to the user level program to do useful things with this 406 information. This is generally a good idea, so say Y. 407 408config BSD_PROCESS_ACCT_V3 409 bool "BSD Process Accounting version 3 file format" 410 depends on BSD_PROCESS_ACCT 411 default n 412 help 413 If you say Y here, the process accounting information is written 414 in a new file format that also logs the process IDs of each 415 process and it's parent. Note that this file format is incompatible 416 with previous v0/v1/v2 file formats, so you will need updated tools 417 for processing it. A preliminary version of these tools is available 418 at <http://www.gnu.org/software/acct/>. 419 420config TASKSTATS 421 bool "Export task/process statistics through netlink" 422 depends on NET 423 default n 424 help 425 Export selected statistics for tasks/processes through the 426 generic netlink interface. Unlike BSD process accounting, the 427 statistics are available during the lifetime of tasks/processes as 428 responses to commands. Like BSD accounting, they are sent to user 429 space on task exit. 430 431 Say N if unsure. 432 433config TASK_DELAY_ACCT 434 bool "Enable per-task delay accounting" 435 depends on TASKSTATS 436 help 437 Collect information on time spent by a task waiting for system 438 resources like cpu, synchronous block I/O completion and swapping 439 in pages. Such statistics can help in setting a task's priorities 440 relative to other tasks for cpu, io, rss limits etc. 441 442 Say N if unsure. 443 444config TASK_XACCT 445 bool "Enable extended accounting over taskstats" 446 depends on TASKSTATS 447 help 448 Collect extended task accounting data and send the data 449 to userland for processing over the taskstats interface. 450 451 Say N if unsure. 452 453config TASK_IO_ACCOUNTING 454 bool "Enable per-task storage I/O accounting" 455 depends on TASK_XACCT 456 help 457 Collect information on the number of bytes of storage I/O which this 458 task has caused. 459 460 Say N if unsure. 461 462endmenu # "CPU/Task time and stats accounting" 463 464menu "RCU Subsystem" 465 466choice 467 prompt "RCU Implementation" 468 default TREE_RCU 469 470config TREE_RCU 471 bool "Tree-based hierarchical RCU" 472 depends on !PREEMPT && SMP 473 select IRQ_WORK 474 help 475 This option selects the RCU implementation that is 476 designed for very large SMP system with hundreds or 477 thousands of CPUs. It also scales down nicely to 478 smaller systems. 479 480config PREEMPT_RCU 481 bool "Preemptible tree-based hierarchical RCU" 482 depends on PREEMPT 483 select IRQ_WORK 484 help 485 This option selects the RCU implementation that is 486 designed for very large SMP systems with hundreds or 487 thousands of CPUs, but for which real-time response 488 is also required. It also scales down nicely to 489 smaller systems. 490 491 Select this option if you are unsure. 492 493config TINY_RCU 494 bool "UP-only small-memory-footprint RCU" 495 depends on !PREEMPT && !SMP 496 help 497 This option selects the RCU implementation that is 498 designed for UP systems from which real-time response 499 is not required. This option greatly reduces the 500 memory footprint of RCU. 501 502endchoice 503 504config TASKS_RCU 505 bool "Task_based RCU implementation using voluntary context switch" 506 default n 507 help 508 This option enables a task-based RCU implementation that uses 509 only voluntary context switch (not preemption!), idle, and 510 user-mode execution as quiescent states. 511 512 If unsure, say N. 513 514config RCU_STALL_COMMON 515 def_bool ( TREE_RCU || PREEMPT_RCU || RCU_TRACE ) 516 help 517 This option enables RCU CPU stall code that is common between 518 the TINY and TREE variants of RCU. The purpose is to allow 519 the tiny variants to disable RCU CPU stall warnings, while 520 making these warnings mandatory for the tree variants. 521 522config CONTEXT_TRACKING 523 bool 524 525config RCU_USER_QS 526 bool "Consider userspace as in RCU extended quiescent state" 527 depends on HAVE_CONTEXT_TRACKING && SMP 528 select CONTEXT_TRACKING 529 help 530 This option sets hooks on kernel / userspace boundaries and 531 puts RCU in extended quiescent state when the CPU runs in 532 userspace. It means that when a CPU runs in userspace, it is 533 excluded from the global RCU state machine and thus doesn't 534 try to keep the timer tick on for RCU. 535 536 Unless you want to hack and help the development of the full 537 dynticks mode, you shouldn't enable this option. It also 538 adds unnecessary overhead. 539 540 If unsure say N 541 542config CONTEXT_TRACKING_FORCE 543 bool "Force context tracking" 544 depends on CONTEXT_TRACKING 545 default y if !NO_HZ_FULL 546 help 547 The major pre-requirement for full dynticks to work is to 548 support the context tracking subsystem. But there are also 549 other dependencies to provide in order to make the full 550 dynticks working. 551 552 This option stands for testing when an arch implements the 553 context tracking backend but doesn't yet fullfill all the 554 requirements to make the full dynticks feature working. 555 Without the full dynticks, there is no way to test the support 556 for context tracking and the subsystems that rely on it: RCU 557 userspace extended quiescent state and tickless cputime 558 accounting. This option copes with the absence of the full 559 dynticks subsystem by forcing the context tracking on all 560 CPUs in the system. 561 562 Say Y only if you're working on the development of an 563 architecture backend for the context tracking. 564 565 Say N otherwise, this option brings an overhead that you 566 don't want in production. 567 568 569config RCU_FANOUT 570 int "Tree-based hierarchical RCU fanout value" 571 range 2 64 if 64BIT 572 range 2 32 if !64BIT 573 depends on TREE_RCU || PREEMPT_RCU 574 default 64 if 64BIT 575 default 32 if !64BIT 576 help 577 This option controls the fanout of hierarchical implementations 578 of RCU, allowing RCU to work efficiently on machines with 579 large numbers of CPUs. This value must be at least the fourth 580 root of NR_CPUS, which allows NR_CPUS to be insanely large. 581 The default value of RCU_FANOUT should be used for production 582 systems, but if you are stress-testing the RCU implementation 583 itself, small RCU_FANOUT values allow you to test large-system 584 code paths on small(er) systems. 585 586 Select a specific number if testing RCU itself. 587 Take the default if unsure. 588 589config RCU_FANOUT_LEAF 590 int "Tree-based hierarchical RCU leaf-level fanout value" 591 range 2 RCU_FANOUT if 64BIT 592 range 2 RCU_FANOUT if !64BIT 593 depends on TREE_RCU || PREEMPT_RCU 594 default 16 595 help 596 This option controls the leaf-level fanout of hierarchical 597 implementations of RCU, and allows trading off cache misses 598 against lock contention. Systems that synchronize their 599 scheduling-clock interrupts for energy-efficiency reasons will 600 want the default because the smaller leaf-level fanout keeps 601 lock contention levels acceptably low. Very large systems 602 (hundreds or thousands of CPUs) will instead want to set this 603 value to the maximum value possible in order to reduce the 604 number of cache misses incurred during RCU's grace-period 605 initialization. These systems tend to run CPU-bound, and thus 606 are not helped by synchronized interrupts, and thus tend to 607 skew them, which reduces lock contention enough that large 608 leaf-level fanouts work well. 609 610 Select a specific number if testing RCU itself. 611 612 Select the maximum permissible value for large systems. 613 614 Take the default if unsure. 615 616config RCU_FANOUT_EXACT 617 bool "Disable tree-based hierarchical RCU auto-balancing" 618 depends on TREE_RCU || PREEMPT_RCU 619 default n 620 help 621 This option forces use of the exact RCU_FANOUT value specified, 622 regardless of imbalances in the hierarchy. This is useful for 623 testing RCU itself, and might one day be useful on systems with 624 strong NUMA behavior. 625 626 Without RCU_FANOUT_EXACT, the code will balance the hierarchy. 627 628 Say N if unsure. 629 630config RCU_FAST_NO_HZ 631 bool "Accelerate last non-dyntick-idle CPU's grace periods" 632 depends on NO_HZ_COMMON && SMP 633 default n 634 help 635 This option permits CPUs to enter dynticks-idle state even if 636 they have RCU callbacks queued, and prevents RCU from waking 637 these CPUs up more than roughly once every four jiffies (by 638 default, you can adjust this using the rcutree.rcu_idle_gp_delay 639 parameter), thus improving energy efficiency. On the other 640 hand, this option increases the duration of RCU grace periods, 641 for example, slowing down synchronize_rcu(). 642 643 Say Y if energy efficiency is critically important, and you 644 don't care about increased grace-period durations. 645 646 Say N if you are unsure. 647 648config TREE_RCU_TRACE 649 def_bool RCU_TRACE && ( TREE_RCU || PREEMPT_RCU ) 650 select DEBUG_FS 651 help 652 This option provides tracing for the TREE_RCU and 653 PREEMPT_RCU implementations, permitting Makefile to 654 trivially select kernel/rcutree_trace.c. 655 656config RCU_BOOST 657 bool "Enable RCU priority boosting" 658 depends on RT_MUTEXES && PREEMPT_RCU 659 default n 660 help 661 This option boosts the priority of preempted RCU readers that 662 block the current preemptible RCU grace period for too long. 663 This option also prevents heavy loads from blocking RCU 664 callback invocation for all flavors of RCU. 665 666 Say Y here if you are working with real-time apps or heavy loads 667 Say N here if you are unsure. 668 669config RCU_KTHREAD_PRIO 670 int "Real-time priority to use for RCU worker threads" 671 range 1 99 672 depends on RCU_BOOST 673 default 1 674 help 675 This option specifies the SCHED_FIFO priority value that will be 676 assigned to the rcuc/n and rcub/n threads and is also the value 677 used for RCU_BOOST (if enabled). If you are working with a 678 real-time application that has one or more CPU-bound threads 679 running at a real-time priority level, you should set 680 RCU_KTHREAD_PRIO to a priority higher than the highest-priority 681 real-time CPU-bound application thread. The default RCU_KTHREAD_PRIO 682 value of 1 is appropriate in the common case, which is real-time 683 applications that do not have any CPU-bound threads. 684 685 Some real-time applications might not have a single real-time 686 thread that saturates a given CPU, but instead might have 687 multiple real-time threads that, taken together, fully utilize 688 that CPU. In this case, you should set RCU_KTHREAD_PRIO to 689 a priority higher than the lowest-priority thread that is 690 conspiring to prevent the CPU from running any non-real-time 691 tasks. For example, if one thread at priority 10 and another 692 thread at priority 5 are between themselves fully consuming 693 the CPU time on a given CPU, then RCU_KTHREAD_PRIO should be 694 set to priority 6 or higher. 695 696 Specify the real-time priority, or take the default if unsure. 697 698config RCU_BOOST_DELAY 699 int "Milliseconds to delay boosting after RCU grace-period start" 700 range 0 3000 701 depends on RCU_BOOST 702 default 500 703 help 704 This option specifies the time to wait after the beginning of 705 a given grace period before priority-boosting preempted RCU 706 readers blocking that grace period. Note that any RCU reader 707 blocking an expedited RCU grace period is boosted immediately. 708 709 Accept the default if unsure. 710 711config RCU_NOCB_CPU 712 bool "Offload RCU callback processing from boot-selected CPUs" 713 depends on TREE_RCU || PREEMPT_RCU 714 default n 715 help 716 Use this option to reduce OS jitter for aggressive HPC or 717 real-time workloads. It can also be used to offload RCU 718 callback invocation to energy-efficient CPUs in battery-powered 719 asymmetric multiprocessors. 720 721 This option offloads callback invocation from the set of 722 CPUs specified at boot time by the rcu_nocbs parameter. 723 For each such CPU, a kthread ("rcuox/N") will be created to 724 invoke callbacks, where the "N" is the CPU being offloaded, 725 and where the "x" is "b" for RCU-bh, "p" for RCU-preempt, and 726 "s" for RCU-sched. Nothing prevents this kthread from running 727 on the specified CPUs, but (1) the kthreads may be preempted 728 between each callback, and (2) affinity or cgroups can be used 729 to force the kthreads to run on whatever set of CPUs is desired. 730 731 Say Y here if you want to help to debug reduced OS jitter. 732 Say N here if you are unsure. 733 734choice 735 prompt "Build-forced no-CBs CPUs" 736 default RCU_NOCB_CPU_NONE 737 depends on RCU_NOCB_CPU 738 help 739 This option allows no-CBs CPUs (whose RCU callbacks are invoked 740 from kthreads rather than from softirq context) to be specified 741 at build time. Additional no-CBs CPUs may be specified by 742 the rcu_nocbs= boot parameter. 743 744config RCU_NOCB_CPU_NONE 745 bool "No build_forced no-CBs CPUs" 746 help 747 This option does not force any of the CPUs to be no-CBs CPUs. 748 Only CPUs designated by the rcu_nocbs= boot parameter will be 749 no-CBs CPUs, whose RCU callbacks will be invoked by per-CPU 750 kthreads whose names begin with "rcuo". All other CPUs will 751 invoke their own RCU callbacks in softirq context. 752 753 Select this option if you want to choose no-CBs CPUs at 754 boot time, for example, to allow testing of different no-CBs 755 configurations without having to rebuild the kernel each time. 756 757config RCU_NOCB_CPU_ZERO 758 bool "CPU 0 is a build_forced no-CBs CPU" 759 help 760 This option forces CPU 0 to be a no-CBs CPU, so that its RCU 761 callbacks are invoked by a per-CPU kthread whose name begins 762 with "rcuo". Additional CPUs may be designated as no-CBs 763 CPUs using the rcu_nocbs= boot parameter will be no-CBs CPUs. 764 All other CPUs will invoke their own RCU callbacks in softirq 765 context. 766 767 Select this if CPU 0 needs to be a no-CBs CPU for real-time 768 or energy-efficiency reasons, but the real reason it exists 769 is to ensure that randconfig testing covers mixed systems. 770 771config RCU_NOCB_CPU_ALL 772 bool "All CPUs are build_forced no-CBs CPUs" 773 help 774 This option forces all CPUs to be no-CBs CPUs. The rcu_nocbs= 775 boot parameter will be ignored. All CPUs' RCU callbacks will 776 be executed in the context of per-CPU rcuo kthreads created for 777 this purpose. Assuming that the kthreads whose names start with 778 "rcuo" are bound to "housekeeping" CPUs, this reduces OS jitter 779 on the remaining CPUs, but might decrease memory locality during 780 RCU-callback invocation, thus potentially degrading throughput. 781 782 Select this if all CPUs need to be no-CBs CPUs for real-time 783 or energy-efficiency reasons. 784 785endchoice 786 787endmenu # "RCU Subsystem" 788 789config BUILD_BIN2C 790 bool 791 default n 792 793config IKCONFIG 794 tristate "Kernel .config support" 795 select BUILD_BIN2C 796 ---help--- 797 This option enables the complete Linux kernel ".config" file 798 contents to be saved in the kernel. It provides documentation 799 of which kernel options are used in a running kernel or in an 800 on-disk kernel. This information can be extracted from the kernel 801 image file with the script scripts/extract-ikconfig and used as 802 input to rebuild the current kernel or to build another kernel. 803 It can also be extracted from a running kernel by reading 804 /proc/config.gz if enabled (below). 805 806config IKCONFIG_PROC 807 bool "Enable access to .config through /proc/config.gz" 808 depends on IKCONFIG && PROC_FS 809 ---help--- 810 This option enables access to the kernel configuration file 811 through /proc/config.gz. 812 813config LOG_BUF_SHIFT 814 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" 815 range 12 21 816 default 17 817 depends on PRINTK 818 help 819 Select the minimal kernel log buffer size as a power of 2. 820 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config 821 parameter, see below. Any higher size also might be forced 822 by "log_buf_len" boot parameter. 823 824 Examples: 825 17 => 128 KB 826 16 => 64 KB 827 15 => 32 KB 828 14 => 16 KB 829 13 => 8 KB 830 12 => 4 KB 831 832config LOG_CPU_MAX_BUF_SHIFT 833 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)" 834 depends on SMP 835 range 0 21 836 default 12 if !BASE_SMALL 837 default 0 if BASE_SMALL 838 depends on PRINTK 839 help 840 This option allows to increase the default ring buffer size 841 according to the number of CPUs. The value defines the contribution 842 of each CPU as a power of 2. The used space is typically only few 843 lines however it might be much more when problems are reported, 844 e.g. backtraces. 845 846 The increased size means that a new buffer has to be allocated and 847 the original static one is unused. It makes sense only on systems 848 with more CPUs. Therefore this value is used only when the sum of 849 contributions is greater than the half of the default kernel ring 850 buffer as defined by LOG_BUF_SHIFT. The default values are set 851 so that more than 64 CPUs are needed to trigger the allocation. 852 853 Also this option is ignored when "log_buf_len" kernel parameter is 854 used as it forces an exact (power of two) size of the ring buffer. 855 856 The number of possible CPUs is used for this computation ignoring 857 hotplugging making the compuation optimal for the the worst case 858 scenerio while allowing a simple algorithm to be used from bootup. 859 860 Examples shift values and their meaning: 861 17 => 128 KB for each CPU 862 16 => 64 KB for each CPU 863 15 => 32 KB for each CPU 864 14 => 16 KB for each CPU 865 13 => 8 KB for each CPU 866 12 => 4 KB for each CPU 867 868# 869# Architectures with an unreliable sched_clock() should select this: 870# 871config HAVE_UNSTABLE_SCHED_CLOCK 872 bool 873 874config GENERIC_SCHED_CLOCK 875 bool 876 877# 878# For architectures that want to enable the support for NUMA-affine scheduler 879# balancing logic: 880# 881config ARCH_SUPPORTS_NUMA_BALANCING 882 bool 883 884# 885# For architectures that know their GCC __int128 support is sound 886# 887config ARCH_SUPPORTS_INT128 888 bool 889 890# For architectures that (ab)use NUMA to represent different memory regions 891# all cpu-local but of different latencies, such as SuperH. 892# 893config ARCH_WANT_NUMA_VARIABLE_LOCALITY 894 bool 895 896config NUMA_BALANCING_DEFAULT_ENABLED 897 bool "Automatically enable NUMA aware memory/task placement" 898 default y 899 depends on NUMA_BALANCING 900 help 901 If set, automatic NUMA balancing will be enabled if running on a NUMA 902 machine. 903 904config NUMA_BALANCING 905 bool "Memory placement aware NUMA scheduler" 906 depends on ARCH_SUPPORTS_NUMA_BALANCING 907 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY 908 depends on SMP && NUMA && MIGRATION 909 help 910 This option adds support for automatic NUMA aware memory/task placement. 911 The mechanism is quite primitive and is based on migrating memory when 912 it has references to the node the task is running on. 913 914 This system will be inactive on UMA systems. 915 916menuconfig CGROUPS 917 boolean "Control Group support" 918 select KERNFS 919 help 920 This option adds support for grouping sets of processes together, for 921 use with process control subsystems such as Cpusets, CFS, memory 922 controls or device isolation. 923 See 924 - Documentation/scheduler/sched-design-CFS.txt (CFS) 925 - Documentation/cgroups/ (features for grouping, isolation 926 and resource control) 927 928 Say N if unsure. 929 930if CGROUPS 931 932config CGROUP_DEBUG 933 bool "Example debug cgroup subsystem" 934 default n 935 help 936 This option enables a simple cgroup subsystem that 937 exports useful debugging information about the cgroups 938 framework. 939 940 Say N if unsure. 941 942config CGROUP_FREEZER 943 bool "Freezer cgroup subsystem" 944 help 945 Provides a way to freeze and unfreeze all tasks in a 946 cgroup. 947 948config CGROUP_DEVICE 949 bool "Device controller for cgroups" 950 help 951 Provides a cgroup implementing whitelists for devices which 952 a process in the cgroup can mknod or open. 953 954config CPUSETS 955 bool "Cpuset support" 956 help 957 This option will let you create and manage CPUSETs which 958 allow dynamically partitioning a system into sets of CPUs and 959 Memory Nodes and assigning tasks to run only within those sets. 960 This is primarily useful on large SMP or NUMA systems. 961 962 Say N if unsure. 963 964config PROC_PID_CPUSET 965 bool "Include legacy /proc/<pid>/cpuset file" 966 depends on CPUSETS 967 default y 968 969config CGROUP_CPUACCT 970 bool "Simple CPU accounting cgroup subsystem" 971 help 972 Provides a simple Resource Controller for monitoring the 973 total CPU consumed by the tasks in a cgroup. 974 975config RESOURCE_COUNTERS 976 bool "Resource counters" 977 help 978 This option enables controller independent resource accounting 979 infrastructure that works with cgroups. 980 981config MEMCG 982 bool "Memory Resource Controller for Control Groups" 983 depends on RESOURCE_COUNTERS 984 select EVENTFD 985 help 986 Provides a memory resource controller that manages both anonymous 987 memory and page cache. (See Documentation/cgroups/memory.txt) 988 989 Note that setting this option increases fixed memory overhead 990 associated with each page of memory in the system. By this, 991 8(16)bytes/PAGE_SIZE on 32(64)bit system will be occupied by memory 992 usage tracking struct at boot. Total amount of this is printed out 993 at boot. 994 995 Only enable when you're ok with these trade offs and really 996 sure you need the memory resource controller. Even when you enable 997 this, you can set "cgroup_disable=memory" at your boot option to 998 disable memory resource controller and you can avoid overheads. 999 (and lose benefits of memory resource controller) 1000 1001config MEMCG_SWAP 1002 bool "Memory Resource Controller Swap Extension" 1003 depends on MEMCG && SWAP 1004 help 1005 Add swap management feature to memory resource controller. When you 1006 enable this, you can limit mem+swap usage per cgroup. In other words, 1007 when you disable this, memory resource controller has no cares to 1008 usage of swap...a process can exhaust all of the swap. This extension 1009 is useful when you want to avoid exhaustion swap but this itself 1010 adds more overheads and consumes memory for remembering information. 1011 Especially if you use 32bit system or small memory system, please 1012 be careful about enabling this. When memory resource controller 1013 is disabled by boot option, this will be automatically disabled and 1014 there will be no overhead from this. Even when you set this config=y, 1015 if boot option "swapaccount=0" is set, swap will not be accounted. 1016 Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page 1017 size is 4096bytes, 512k per 1Gbytes of swap. 1018config MEMCG_SWAP_ENABLED 1019 bool "Memory Resource Controller Swap Extension enabled by default" 1020 depends on MEMCG_SWAP 1021 default y 1022 help 1023 Memory Resource Controller Swap Extension comes with its price in 1024 a bigger memory consumption. General purpose distribution kernels 1025 which want to enable the feature but keep it disabled by default 1026 and let the user enable it by swapaccount=1 boot command line 1027 parameter should have this option unselected. 1028 For those who want to have the feature enabled by default should 1029 select this option (if, for some reason, they need to disable it 1030 then swapaccount=0 does the trick). 1031config MEMCG_KMEM 1032 bool "Memory Resource Controller Kernel Memory accounting" 1033 depends on MEMCG 1034 depends on SLUB || SLAB 1035 help 1036 The Kernel Memory extension for Memory Resource Controller can limit 1037 the amount of memory used by kernel objects in the system. Those are 1038 fundamentally different from the entities handled by the standard 1039 Memory Controller, which are page-based, and can be swapped. Users of 1040 the kmem extension can use it to guarantee that no group of processes 1041 will ever exhaust kernel resources alone. 1042 1043 WARNING: Current implementation lacks reclaim support. That means 1044 allocation attempts will fail when close to the limit even if there 1045 are plenty of kmem available for reclaim. That makes this option 1046 unusable in real life so DO NOT SELECT IT unless for development 1047 purposes. 1048 1049config CGROUP_HUGETLB 1050 bool "HugeTLB Resource Controller for Control Groups" 1051 depends on RESOURCE_COUNTERS && HUGETLB_PAGE 1052 default n 1053 help 1054 Provides a cgroup Resource Controller for HugeTLB pages. 1055 When you enable this, you can put a per cgroup limit on HugeTLB usage. 1056 The limit is enforced during page fault. Since HugeTLB doesn't 1057 support page reclaim, enforcing the limit at page fault time implies 1058 that, the application will get SIGBUS signal if it tries to access 1059 HugeTLB pages beyond its limit. This requires the application to know 1060 beforehand how much HugeTLB pages it would require for its use. The 1061 control group is tracked in the third page lru pointer. This means 1062 that we cannot use the controller with huge page less than 3 pages. 1063 1064config CGROUP_PERF 1065 bool "Enable perf_event per-cpu per-container group (cgroup) monitoring" 1066 depends on PERF_EVENTS && CGROUPS 1067 help 1068 This option extends the per-cpu mode to restrict monitoring to 1069 threads which belong to the cgroup specified and run on the 1070 designated cpu. 1071 1072 Say N if unsure. 1073 1074menuconfig CGROUP_SCHED 1075 bool "Group CPU scheduler" 1076 default n 1077 help 1078 This feature lets CPU scheduler recognize task groups and control CPU 1079 bandwidth allocation to such task groups. It uses cgroups to group 1080 tasks. 1081 1082if CGROUP_SCHED 1083config FAIR_GROUP_SCHED 1084 bool "Group scheduling for SCHED_OTHER" 1085 depends on CGROUP_SCHED 1086 default CGROUP_SCHED 1087 1088config CFS_BANDWIDTH 1089 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED" 1090 depends on FAIR_GROUP_SCHED 1091 default n 1092 help 1093 This option allows users to define CPU bandwidth rates (limits) for 1094 tasks running within the fair group scheduler. Groups with no limit 1095 set are considered to be unconstrained and will run with no 1096 restriction. 1097 See tip/Documentation/scheduler/sched-bwc.txt for more information. 1098 1099config RT_GROUP_SCHED 1100 bool "Group scheduling for SCHED_RR/FIFO" 1101 depends on CGROUP_SCHED 1102 default n 1103 help 1104 This feature lets you explicitly allocate real CPU bandwidth 1105 to task groups. If enabled, it will also make it impossible to 1106 schedule realtime tasks for non-root users until you allocate 1107 realtime bandwidth for them. 1108 See Documentation/scheduler/sched-rt-group.txt for more information. 1109 1110endif #CGROUP_SCHED 1111 1112config BLK_CGROUP 1113 bool "Block IO controller" 1114 depends on BLOCK 1115 default n 1116 ---help--- 1117 Generic block IO controller cgroup interface. This is the common 1118 cgroup interface which should be used by various IO controlling 1119 policies. 1120 1121 Currently, CFQ IO scheduler uses it to recognize task groups and 1122 control disk bandwidth allocation (proportional time slice allocation) 1123 to such task groups. It is also used by bio throttling logic in 1124 block layer to implement upper limit in IO rates on a device. 1125 1126 This option only enables generic Block IO controller infrastructure. 1127 One needs to also enable actual IO controlling logic/policy. For 1128 enabling proportional weight division of disk bandwidth in CFQ, set 1129 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set 1130 CONFIG_BLK_DEV_THROTTLING=y. 1131 1132 See Documentation/cgroups/blkio-controller.txt for more information. 1133 1134config DEBUG_BLK_CGROUP 1135 bool "Enable Block IO controller debugging" 1136 depends on BLK_CGROUP 1137 default n 1138 ---help--- 1139 Enable some debugging help. Currently it exports additional stat 1140 files in a cgroup which can be useful for debugging. 1141 1142endif # CGROUPS 1143 1144config CHECKPOINT_RESTORE 1145 bool "Checkpoint/restore support" if EXPERT 1146 default n 1147 help 1148 Enables additional kernel features in a sake of checkpoint/restore. 1149 In particular it adds auxiliary prctl codes to setup process text, 1150 data and heap segment sizes, and a few additional /proc filesystem 1151 entries. 1152 1153 If unsure, say N here. 1154 1155menuconfig NAMESPACES 1156 bool "Namespaces support" if EXPERT 1157 default !EXPERT 1158 help 1159 Provides the way to make tasks work with different objects using 1160 the same id. For example same IPC id may refer to different objects 1161 or same user id or pid may refer to different tasks when used in 1162 different namespaces. 1163 1164if NAMESPACES 1165 1166config UTS_NS 1167 bool "UTS namespace" 1168 default y 1169 help 1170 In this namespace tasks see different info provided with the 1171 uname() system call 1172 1173config IPC_NS 1174 bool "IPC namespace" 1175 depends on (SYSVIPC || POSIX_MQUEUE) 1176 default y 1177 help 1178 In this namespace tasks work with IPC ids which correspond to 1179 different IPC objects in different namespaces. 1180 1181config USER_NS 1182 bool "User namespace" 1183 default n 1184 help 1185 This allows containers, i.e. vservers, to use user namespaces 1186 to provide different user info for different servers. 1187 1188 When user namespaces are enabled in the kernel it is 1189 recommended that the MEMCG and MEMCG_KMEM options also be 1190 enabled and that user-space use the memory control groups to 1191 limit the amount of memory a memory unprivileged users can 1192 use. 1193 1194 If unsure, say N. 1195 1196config PID_NS 1197 bool "PID Namespaces" 1198 default y 1199 help 1200 Support process id namespaces. This allows having multiple 1201 processes with the same pid as long as they are in different 1202 pid namespaces. This is a building block of containers. 1203 1204config NET_NS 1205 bool "Network namespace" 1206 depends on NET 1207 default y 1208 help 1209 Allow user space to create what appear to be multiple instances 1210 of the network stack. 1211 1212endif # NAMESPACES 1213 1214config SCHED_AUTOGROUP 1215 bool "Automatic process group scheduling" 1216 select CGROUPS 1217 select CGROUP_SCHED 1218 select FAIR_GROUP_SCHED 1219 help 1220 This option optimizes the scheduler for common desktop workloads by 1221 automatically creating and populating task groups. This separation 1222 of workloads isolates aggressive CPU burners (like build jobs) from 1223 desktop applications. Task group autogeneration is currently based 1224 upon task session. 1225 1226config SYSFS_DEPRECATED 1227 bool "Enable deprecated sysfs features to support old userspace tools" 1228 depends on SYSFS 1229 default n 1230 help 1231 This option adds code that switches the layout of the "block" class 1232 devices, to not show up in /sys/class/block/, but only in 1233 /sys/block/. 1234 1235 This switch is only active when the sysfs.deprecated=1 boot option is 1236 passed or the SYSFS_DEPRECATED_V2 option is set. 1237 1238 This option allows new kernels to run on old distributions and tools, 1239 which might get confused by /sys/class/block/. Since 2007/2008 all 1240 major distributions and tools handle this just fine. 1241 1242 Recent distributions and userspace tools after 2009/2010 depend on 1243 the existence of /sys/class/block/, and will not work with this 1244 option enabled. 1245 1246 Only if you are using a new kernel on an old distribution, you might 1247 need to say Y here. 1248 1249config SYSFS_DEPRECATED_V2 1250 bool "Enable deprecated sysfs features by default" 1251 default n 1252 depends on SYSFS 1253 depends on SYSFS_DEPRECATED 1254 help 1255 Enable deprecated sysfs by default. 1256 1257 See the CONFIG_SYSFS_DEPRECATED option for more details about this 1258 option. 1259 1260 Only if you are using a new kernel on an old distribution, you might 1261 need to say Y here. Even then, odds are you would not need it 1262 enabled, you can always pass the boot option if absolutely necessary. 1263 1264config RELAY 1265 bool "Kernel->user space relay support (formerly relayfs)" 1266 help 1267 This option enables support for relay interface support in 1268 certain file systems (such as debugfs). 1269 It is designed to provide an efficient mechanism for tools and 1270 facilities to relay large amounts of data from kernel space to 1271 user space. 1272 1273 If unsure, say N. 1274 1275config BLK_DEV_INITRD 1276 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support" 1277 depends on BROKEN || !FRV 1278 help 1279 The initial RAM filesystem is a ramfs which is loaded by the 1280 boot loader (loadlin or lilo) and that is mounted as root 1281 before the normal boot procedure. It is typically used to 1282 load modules needed to mount the "real" root file system, 1283 etc. See <file:Documentation/initrd.txt> for details. 1284 1285 If RAM disk support (BLK_DEV_RAM) is also included, this 1286 also enables initial RAM disk (initrd) support and adds 1287 15 Kbytes (more on some other architectures) to the kernel size. 1288 1289 If unsure say Y. 1290 1291if BLK_DEV_INITRD 1292 1293source "usr/Kconfig" 1294 1295endif 1296 1297config CC_OPTIMIZE_FOR_SIZE 1298 bool "Optimize for size" 1299 help 1300 Enabling this option will pass "-Os" instead of "-O2" to gcc 1301 resulting in a smaller kernel. 1302 1303 If unsure, say N. 1304 1305config SYSCTL 1306 bool 1307 1308config ANON_INODES 1309 bool 1310 1311config HAVE_UID16 1312 bool 1313 1314config SYSCTL_EXCEPTION_TRACE 1315 bool 1316 help 1317 Enable support for /proc/sys/debug/exception-trace. 1318 1319config SYSCTL_ARCH_UNALIGN_NO_WARN 1320 bool 1321 help 1322 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap 1323 Allows arch to define/use @no_unaligned_warning to possibly warn 1324 about unaligned access emulation going on under the hood. 1325 1326config SYSCTL_ARCH_UNALIGN_ALLOW 1327 bool 1328 help 1329 Enable support for /proc/sys/kernel/unaligned-trap 1330 Allows arches to define/use @unaligned_enabled to runtime toggle 1331 the unaligned access emulation. 1332 see arch/parisc/kernel/unaligned.c for reference 1333 1334config HAVE_PCSPKR_PLATFORM 1335 bool 1336 1337# interpreter that classic socket filters depend on 1338config BPF 1339 bool 1340 1341menuconfig EXPERT 1342 bool "Configure standard kernel features (expert users)" 1343 # Unhide debug options, to make the on-by-default options visible 1344 select DEBUG_KERNEL 1345 help 1346 This option allows certain base kernel options and settings 1347 to be disabled or tweaked. This is for specialized 1348 environments which can tolerate a "non-standard" kernel. 1349 Only use this if you really know what you are doing. 1350 1351config UID16 1352 bool "Enable 16-bit UID system calls" if EXPERT 1353 depends on HAVE_UID16 1354 default y 1355 help 1356 This enables the legacy 16-bit UID syscall wrappers. 1357 1358config SGETMASK_SYSCALL 1359 bool "sgetmask/ssetmask syscalls support" if EXPERT 1360 def_bool PARISC || MN10300 || BLACKFIN || M68K || PPC || MIPS || X86 || SPARC || CRIS || MICROBLAZE || SUPERH 1361 ---help--- 1362 sys_sgetmask and sys_ssetmask are obsolete system calls 1363 no longer supported in libc but still enabled by default in some 1364 architectures. 1365 1366 If unsure, leave the default option here. 1367 1368config SYSFS_SYSCALL 1369 bool "Sysfs syscall support" if EXPERT 1370 default y 1371 ---help--- 1372 sys_sysfs is an obsolete system call no longer supported in libc. 1373 Note that disabling this option is more secure but might break 1374 compatibility with some systems. 1375 1376 If unsure say Y here. 1377 1378config SYSCTL_SYSCALL 1379 bool "Sysctl syscall support" if EXPERT 1380 depends on PROC_SYSCTL 1381 default n 1382 select SYSCTL 1383 ---help--- 1384 sys_sysctl uses binary paths that have been found challenging 1385 to properly maintain and use. The interface in /proc/sys 1386 using paths with ascii names is now the primary path to this 1387 information. 1388 1389 Almost nothing using the binary sysctl interface so if you are 1390 trying to save some space it is probably safe to disable this, 1391 making your kernel marginally smaller. 1392 1393 If unsure say N here. 1394 1395config KALLSYMS 1396 bool "Load all symbols for debugging/ksymoops" if EXPERT 1397 default y 1398 help 1399 Say Y here to let the kernel print out symbolic crash information and 1400 symbolic stack backtraces. This increases the size of the kernel 1401 somewhat, as all symbols have to be loaded into the kernel image. 1402 1403config KALLSYMS_ALL 1404 bool "Include all symbols in kallsyms" 1405 depends on DEBUG_KERNEL && KALLSYMS 1406 help 1407 Normally kallsyms only contains the symbols of functions for nicer 1408 OOPS messages and backtraces (i.e., symbols from the text and inittext 1409 sections). This is sufficient for most cases. And only in very rare 1410 cases (e.g., when a debugger is used) all symbols are required (e.g., 1411 names of variables from the data sections, etc). 1412 1413 This option makes sure that all symbols are loaded into the kernel 1414 image (i.e., symbols from all sections) in cost of increased kernel 1415 size (depending on the kernel configuration, it may be 300KiB or 1416 something like this). 1417 1418 Say N unless you really need all symbols. 1419 1420config PRINTK 1421 default y 1422 bool "Enable support for printk" if EXPERT 1423 select IRQ_WORK 1424 help 1425 This option enables normal printk support. Removing it 1426 eliminates most of the message strings from the kernel image 1427 and makes the kernel more or less silent. As this makes it 1428 very difficult to diagnose system problems, saying N here is 1429 strongly discouraged. 1430 1431config BUG 1432 bool "BUG() support" if EXPERT 1433 default y 1434 help 1435 Disabling this option eliminates support for BUG and WARN, reducing 1436 the size of your kernel image and potentially quietly ignoring 1437 numerous fatal conditions. You should only consider disabling this 1438 option for embedded systems with no facilities for reporting errors. 1439 Just say Y. 1440 1441config ELF_CORE 1442 depends on COREDUMP 1443 default y 1444 bool "Enable ELF core dumps" if EXPERT 1445 help 1446 Enable support for generating core dumps. Disabling saves about 4k. 1447 1448 1449config PCSPKR_PLATFORM 1450 bool "Enable PC-Speaker support" if EXPERT 1451 depends on HAVE_PCSPKR_PLATFORM 1452 select I8253_LOCK 1453 default y 1454 help 1455 This option allows to disable the internal PC-Speaker 1456 support, saving some memory. 1457 1458config BASE_FULL 1459 default y 1460 bool "Enable full-sized data structures for core" if EXPERT 1461 help 1462 Disabling this option reduces the size of miscellaneous core 1463 kernel data structures. This saves memory on small machines, 1464 but may reduce performance. 1465 1466config FUTEX 1467 bool "Enable futex support" if EXPERT 1468 default y 1469 select RT_MUTEXES 1470 help 1471 Disabling this option will cause the kernel to be built without 1472 support for "fast userspace mutexes". The resulting kernel may not 1473 run glibc-based applications correctly. 1474 1475config HAVE_FUTEX_CMPXCHG 1476 bool 1477 depends on FUTEX 1478 help 1479 Architectures should select this if futex_atomic_cmpxchg_inatomic() 1480 is implemented and always working. This removes a couple of runtime 1481 checks. 1482 1483config EPOLL 1484 bool "Enable eventpoll support" if EXPERT 1485 default y 1486 select ANON_INODES 1487 help 1488 Disabling this option will cause the kernel to be built without 1489 support for epoll family of system calls. 1490 1491config SIGNALFD 1492 bool "Enable signalfd() system call" if EXPERT 1493 select ANON_INODES 1494 default y 1495 help 1496 Enable the signalfd() system call that allows to receive signals 1497 on a file descriptor. 1498 1499 If unsure, say Y. 1500 1501config TIMERFD 1502 bool "Enable timerfd() system call" if EXPERT 1503 select ANON_INODES 1504 default y 1505 help 1506 Enable the timerfd() system call that allows to receive timer 1507 events on a file descriptor. 1508 1509 If unsure, say Y. 1510 1511config EVENTFD 1512 bool "Enable eventfd() system call" if EXPERT 1513 select ANON_INODES 1514 default y 1515 help 1516 Enable the eventfd() system call that allows to receive both 1517 kernel notification (ie. KAIO) or userspace notifications. 1518 1519 If unsure, say Y. 1520 1521# syscall, maps, verifier 1522config BPF_SYSCALL 1523 bool "Enable bpf() system call" if EXPERT 1524 select ANON_INODES 1525 select BPF 1526 default n 1527 help 1528 Enable the bpf() system call that allows to manipulate eBPF 1529 programs and maps via file descriptors. 1530 1531config SHMEM 1532 bool "Use full shmem filesystem" if EXPERT 1533 default y 1534 depends on MMU 1535 help 1536 The shmem is an internal filesystem used to manage shared memory. 1537 It is backed by swap and manages resource limits. It is also exported 1538 to userspace as tmpfs if TMPFS is enabled. Disabling this 1539 option replaces shmem and tmpfs with the much simpler ramfs code, 1540 which may be appropriate on small systems without swap. 1541 1542config AIO 1543 bool "Enable AIO support" if EXPERT 1544 default y 1545 help 1546 This option enables POSIX asynchronous I/O which may by used 1547 by some high performance threaded applications. Disabling 1548 this option saves about 7k. 1549 1550config ADVISE_SYSCALLS 1551 bool "Enable madvise/fadvise syscalls" if EXPERT 1552 default y 1553 help 1554 This option enables the madvise and fadvise syscalls, used by 1555 applications to advise the kernel about their future memory or file 1556 usage, improving performance. If building an embedded system where no 1557 applications use these syscalls, you can disable this option to save 1558 space. 1559 1560config PCI_QUIRKS 1561 default y 1562 bool "Enable PCI quirk workarounds" if EXPERT 1563 depends on PCI 1564 help 1565 This enables workarounds for various PCI chipset 1566 bugs/quirks. Disable this only if your target machine is 1567 unaffected by PCI quirks. 1568 1569config EMBEDDED 1570 bool "Embedded system" 1571 option allnoconfig_y 1572 select EXPERT 1573 help 1574 This option should be enabled if compiling the kernel for 1575 an embedded system so certain expert options are available 1576 for configuration. 1577 1578config HAVE_PERF_EVENTS 1579 bool 1580 help 1581 See tools/perf/design.txt for details. 1582 1583config PERF_USE_VMALLOC 1584 bool 1585 help 1586 See tools/perf/design.txt for details 1587 1588menu "Kernel Performance Events And Counters" 1589 1590config PERF_EVENTS 1591 bool "Kernel performance events and counters" 1592 default y if PROFILING 1593 depends on HAVE_PERF_EVENTS 1594 select ANON_INODES 1595 select IRQ_WORK 1596 help 1597 Enable kernel support for various performance events provided 1598 by software and hardware. 1599 1600 Software events are supported either built-in or via the 1601 use of generic tracepoints. 1602 1603 Most modern CPUs support performance events via performance 1604 counter registers. These registers count the number of certain 1605 types of hw events: such as instructions executed, cachemisses 1606 suffered, or branches mis-predicted - without slowing down the 1607 kernel or applications. These registers can also trigger interrupts 1608 when a threshold number of events have passed - and can thus be 1609 used to profile the code that runs on that CPU. 1610 1611 The Linux Performance Event subsystem provides an abstraction of 1612 these software and hardware event capabilities, available via a 1613 system call and used by the "perf" utility in tools/perf/. It 1614 provides per task and per CPU counters, and it provides event 1615 capabilities on top of those. 1616 1617 Say Y if unsure. 1618 1619config DEBUG_PERF_USE_VMALLOC 1620 default n 1621 bool "Debug: use vmalloc to back perf mmap() buffers" 1622 depends on PERF_EVENTS && DEBUG_KERNEL 1623 select PERF_USE_VMALLOC 1624 help 1625 Use vmalloc memory to back perf mmap() buffers. 1626 1627 Mostly useful for debugging the vmalloc code on platforms 1628 that don't require it. 1629 1630 Say N if unsure. 1631 1632endmenu 1633 1634config VM_EVENT_COUNTERS 1635 default y 1636 bool "Enable VM event counters for /proc/vmstat" if EXPERT 1637 help 1638 VM event counters are needed for event counts to be shown. 1639 This option allows the disabling of the VM event counters 1640 on EXPERT systems. /proc/vmstat will only show page counts 1641 if VM event counters are disabled. 1642 1643config SLUB_DEBUG 1644 default y 1645 bool "Enable SLUB debugging support" if EXPERT 1646 depends on SLUB && SYSFS 1647 help 1648 SLUB has extensive debug support features. Disabling these can 1649 result in significant savings in code size. This also disables 1650 SLUB sysfs support. /sys/slab will not exist and there will be 1651 no support for cache validation etc. 1652 1653config COMPAT_BRK 1654 bool "Disable heap randomization" 1655 default y 1656 help 1657 Randomizing heap placement makes heap exploits harder, but it 1658 also breaks ancient binaries (including anything libc5 based). 1659 This option changes the bootup default to heap randomization 1660 disabled, and can be overridden at runtime by setting 1661 /proc/sys/kernel/randomize_va_space to 2. 1662 1663 On non-ancient distros (post-2000 ones) N is usually a safe choice. 1664 1665choice 1666 prompt "Choose SLAB allocator" 1667 default SLUB 1668 help 1669 This option allows to select a slab allocator. 1670 1671config SLAB 1672 bool "SLAB" 1673 help 1674 The regular slab allocator that is established and known to work 1675 well in all environments. It organizes cache hot objects in 1676 per cpu and per node queues. 1677 1678config SLUB 1679 bool "SLUB (Unqueued Allocator)" 1680 help 1681 SLUB is a slab allocator that minimizes cache line usage 1682 instead of managing queues of cached objects (SLAB approach). 1683 Per cpu caching is realized using slabs of objects instead 1684 of queues of objects. SLUB can use memory efficiently 1685 and has enhanced diagnostics. SLUB is the default choice for 1686 a slab allocator. 1687 1688config SLOB 1689 depends on EXPERT 1690 bool "SLOB (Simple Allocator)" 1691 help 1692 SLOB replaces the stock allocator with a drastically simpler 1693 allocator. SLOB is generally more space efficient but 1694 does not perform as well on large systems. 1695 1696endchoice 1697 1698config SLUB_CPU_PARTIAL 1699 default y 1700 depends on SLUB && SMP 1701 bool "SLUB per cpu partial cache" 1702 help 1703 Per cpu partial caches accellerate objects allocation and freeing 1704 that is local to a processor at the price of more indeterminism 1705 in the latency of the free. On overflow these caches will be cleared 1706 which requires the taking of locks that may cause latency spikes. 1707 Typically one would choose no for a realtime system. 1708 1709config MMAP_ALLOW_UNINITIALIZED 1710 bool "Allow mmapped anonymous memory to be uninitialized" 1711 depends on EXPERT && !MMU 1712 default n 1713 help 1714 Normally, and according to the Linux spec, anonymous memory obtained 1715 from mmap() has it's contents cleared before it is passed to 1716 userspace. Enabling this config option allows you to request that 1717 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus 1718 providing a huge performance boost. If this option is not enabled, 1719 then the flag will be ignored. 1720 1721 This is taken advantage of by uClibc's malloc(), and also by 1722 ELF-FDPIC binfmt's brk and stack allocator. 1723 1724 Because of the obvious security issues, this option should only be 1725 enabled on embedded devices where you control what is run in 1726 userspace. Since that isn't generally a problem on no-MMU systems, 1727 it is normally safe to say Y here. 1728 1729 See Documentation/nommu-mmap.txt for more information. 1730 1731config SYSTEM_TRUSTED_KEYRING 1732 bool "Provide system-wide ring of trusted keys" 1733 depends on KEYS 1734 help 1735 Provide a system keyring to which trusted keys can be added. Keys in 1736 the keyring are considered to be trusted. Keys may be added at will 1737 by the kernel from compiled-in data and from hardware key stores, but 1738 userspace may only add extra keys if those keys can be verified by 1739 keys already in the keyring. 1740 1741 Keys in this keyring are used by module signature checking. 1742 1743config PROFILING 1744 bool "Profiling support" 1745 help 1746 Say Y here to enable the extended profiling support mechanisms used 1747 by profilers such as OProfile. 1748 1749# 1750# Place an empty function call at each tracepoint site. Can be 1751# dynamically changed for a probe function. 1752# 1753config TRACEPOINTS 1754 bool 1755 1756source "arch/Kconfig" 1757 1758endmenu # General setup 1759 1760config HAVE_GENERIC_DMA_COHERENT 1761 bool 1762 default n 1763 1764config SLABINFO 1765 bool 1766 depends on PROC_FS 1767 depends on SLAB || SLUB_DEBUG 1768 default y 1769 1770config RT_MUTEXES 1771 boolean 1772 1773config BASE_SMALL 1774 int 1775 default 0 if BASE_FULL 1776 default 1 if !BASE_FULL 1777 1778menuconfig MODULES 1779 bool "Enable loadable module support" 1780 option modules 1781 help 1782 Kernel modules are small pieces of compiled code which can 1783 be inserted in the running kernel, rather than being 1784 permanently built into the kernel. You use the "modprobe" 1785 tool to add (and sometimes remove) them. If you say Y here, 1786 many parts of the kernel can be built as modules (by 1787 answering M instead of Y where indicated): this is most 1788 useful for infrequently used options which are not required 1789 for booting. For more information, see the man pages for 1790 modprobe, lsmod, modinfo, insmod and rmmod. 1791 1792 If you say Y here, you will need to run "make 1793 modules_install" to put the modules under /lib/modules/ 1794 where modprobe can find them (you may need to be root to do 1795 this). 1796 1797 If unsure, say Y. 1798 1799if MODULES 1800 1801config MODULE_FORCE_LOAD 1802 bool "Forced module loading" 1803 default n 1804 help 1805 Allow loading of modules without version information (ie. modprobe 1806 --force). Forced module loading sets the 'F' (forced) taint flag and 1807 is usually a really bad idea. 1808 1809config MODULE_UNLOAD 1810 bool "Module unloading" 1811 help 1812 Without this option you will not be able to unload any 1813 modules (note that some modules may not be unloadable 1814 anyway), which makes your kernel smaller, faster 1815 and simpler. If unsure, say Y. 1816 1817config MODULE_FORCE_UNLOAD 1818 bool "Forced module unloading" 1819 depends on MODULE_UNLOAD 1820 help 1821 This option allows you to force a module to unload, even if the 1822 kernel believes it is unsafe: the kernel will remove the module 1823 without waiting for anyone to stop using it (using the -f option to 1824 rmmod). This is mainly for kernel developers and desperate users. 1825 If unsure, say N. 1826 1827config MODVERSIONS 1828 bool "Module versioning support" 1829 help 1830 Usually, you have to use modules compiled with your kernel. 1831 Saying Y here makes it sometimes possible to use modules 1832 compiled for different kernels, by adding enough information 1833 to the modules to (hopefully) spot any changes which would 1834 make them incompatible with the kernel you are running. If 1835 unsure, say N. 1836 1837config MODULE_SRCVERSION_ALL 1838 bool "Source checksum for all modules" 1839 help 1840 Modules which contain a MODULE_VERSION get an extra "srcversion" 1841 field inserted into their modinfo section, which contains a 1842 sum of the source files which made it. This helps maintainers 1843 see exactly which source was used to build a module (since 1844 others sometimes change the module source without updating 1845 the version). With this option, such a "srcversion" field 1846 will be created for all modules. If unsure, say N. 1847 1848config MODULE_SIG 1849 bool "Module signature verification" 1850 depends on MODULES 1851 select SYSTEM_TRUSTED_KEYRING 1852 select KEYS 1853 select CRYPTO 1854 select ASYMMETRIC_KEY_TYPE 1855 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE 1856 select PUBLIC_KEY_ALGO_RSA 1857 select ASN1 1858 select OID_REGISTRY 1859 select X509_CERTIFICATE_PARSER 1860 help 1861 Check modules for valid signatures upon load: the signature 1862 is simply appended to the module. For more information see 1863 Documentation/module-signing.txt. 1864 1865 !!!WARNING!!! If you enable this option, you MUST make sure that the 1866 module DOES NOT get stripped after being signed. This includes the 1867 debuginfo strip done by some packagers (such as rpmbuild) and 1868 inclusion into an initramfs that wants the module size reduced. 1869 1870config MODULE_SIG_FORCE 1871 bool "Require modules to be validly signed" 1872 depends on MODULE_SIG 1873 help 1874 Reject unsigned modules or signed modules for which we don't have a 1875 key. Without this, such modules will simply taint the kernel. 1876 1877config MODULE_SIG_ALL 1878 bool "Automatically sign all modules" 1879 default y 1880 depends on MODULE_SIG 1881 help 1882 Sign all modules during make modules_install. Without this option, 1883 modules must be signed manually, using the scripts/sign-file tool. 1884 1885comment "Do not forget to sign required modules with scripts/sign-file" 1886 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL 1887 1888choice 1889 prompt "Which hash algorithm should modules be signed with?" 1890 depends on MODULE_SIG 1891 help 1892 This determines which sort of hashing algorithm will be used during 1893 signature generation. This algorithm _must_ be built into the kernel 1894 directly so that signature verification can take place. It is not 1895 possible to load a signed module containing the algorithm to check 1896 the signature on that module. 1897 1898config MODULE_SIG_SHA1 1899 bool "Sign modules with SHA-1" 1900 select CRYPTO_SHA1 1901 1902config MODULE_SIG_SHA224 1903 bool "Sign modules with SHA-224" 1904 select CRYPTO_SHA256 1905 1906config MODULE_SIG_SHA256 1907 bool "Sign modules with SHA-256" 1908 select CRYPTO_SHA256 1909 1910config MODULE_SIG_SHA384 1911 bool "Sign modules with SHA-384" 1912 select CRYPTO_SHA512 1913 1914config MODULE_SIG_SHA512 1915 bool "Sign modules with SHA-512" 1916 select CRYPTO_SHA512 1917 1918endchoice 1919 1920config MODULE_SIG_HASH 1921 string 1922 depends on MODULE_SIG 1923 default "sha1" if MODULE_SIG_SHA1 1924 default "sha224" if MODULE_SIG_SHA224 1925 default "sha256" if MODULE_SIG_SHA256 1926 default "sha384" if MODULE_SIG_SHA384 1927 default "sha512" if MODULE_SIG_SHA512 1928 1929config MODULE_COMPRESS 1930 bool "Compress modules on installation" 1931 depends on MODULES 1932 help 1933 This option compresses the kernel modules when 'make 1934 modules_install' is run. 1935 1936 The modules will be compressed either using gzip or xz depend on the 1937 choice made in "Compression algorithm". 1938 1939 module-init-tools has support for gzip format while kmod handle gzip 1940 and xz compressed modules. 1941 1942 When a kernel module is installed from outside of the main kernel 1943 source and uses the Kbuild system for installing modules then that 1944 kernel module will also be compressed when it is installed. 1945 1946 This option provides little benefit when the modules are to be used inside 1947 an initrd or initramfs, it generally is more efficient to compress the whole 1948 initrd or initramfs instead. 1949 1950 This is fully compatible with signed modules while the signed module is 1951 compressed. module-init-tools or kmod handles decompression and provide to 1952 other layer the uncompressed but signed payload. 1953 1954choice 1955 prompt "Compression algorithm" 1956 depends on MODULE_COMPRESS 1957 default MODULE_COMPRESS_GZIP 1958 help 1959 This determines which sort of compression will be used during 1960 'make modules_install'. 1961 1962 GZIP (default) and XZ are supported. 1963 1964config MODULE_COMPRESS_GZIP 1965 bool "GZIP" 1966 1967config MODULE_COMPRESS_XZ 1968 bool "XZ" 1969 1970endchoice 1971 1972endif # MODULES 1973 1974config INIT_ALL_POSSIBLE 1975 bool 1976 help 1977 Back when each arch used to define their own cpu_online_mask and 1978 cpu_possible_mask, some of them chose to initialize cpu_possible_mask 1979 with all 1s, and others with all 0s. When they were centralised, 1980 it was better to provide this option than to break all the archs 1981 and have several arch maintainers pursuing me down dark alleys. 1982 1983config STOP_MACHINE 1984 bool 1985 default y 1986 depends on (SMP && MODULE_UNLOAD) || HOTPLUG_CPU 1987 help 1988 Need stop_machine() primitive. 1989 1990source "block/Kconfig" 1991 1992config PREEMPT_NOTIFIERS 1993 bool 1994 1995config PADATA 1996 depends on SMP 1997 bool 1998 1999# Can be selected by architectures with broken toolchains 2000# that get confused by correct const<->read_only section 2001# mappings 2002config BROKEN_RODATA 2003 bool 2004 2005config ASN1 2006 tristate 2007 help 2008 Build a simple ASN.1 grammar compiler that produces a bytecode output 2009 that can be interpreted by the ASN.1 stream decoder and used to 2010 inform it as to what tags are to be expected in a stream and what 2011 functions to call on what tags. 2012 2013source "kernel/Kconfig.locks" 2014